By: Dan Sperling, MD
Tissue samples from needle biopsy of the prostate may not yield accurate diagnosis, partly due to sampling error and partly due to misinterpretation by the pathologist who examines the slides. It is not unusual for biopsy results to be at variance with results found on radical prostatectomy (RP). For example, when surgical specimens are analyzed and compared with original biopsy reports, the Gleason score may be upgraded from Gleason 3+3 to Gleason 3+4 or Gleason 4+3. A new Turkish study reported a 40.8% rate of upgrading after RP.[i] Error can also result from overestimation of the Gleason grade. According to one study of 1317 RP patients whose biopsy-proven cancer was graded Gleason 3+4, 115 patients (9%) were in fact determined to have Gleason 3+3 after surgery. In other words, their cancer was downgraded, suggesting they may have been candidates for active surveillance.[ii]
A Memorial Sloan-Kettering research team led by visiting investigator Tatsuo Gondo from Tokyo Medical University retrospectively analyzed the multiparametric MRI (mpMRI) images of 304 patients with biopsy-demonstrated Gleason 3+4 prostate cancer to determine if the images could predict downgrading of the surgical specimens.[iii] All mpMRI imaging was T2 weighted, a basic MRI pulse sequence that reveals zonal anatomy within the prostate as well as differences between cancerous and noncancerous tissues. Additional parameters included diffusion weighted imaging (DWI) and dynamic contrast enhancement (DCE-MRI), each of which can add specific types of information to further identify prostate cancer. The images were then assessed according to the following combinations of parameters:
- T2-weighted imaging alone
- T2-weighted imaging + DWI
- T2-weighted imaging + DCE-MRI
- T2-weighted imaging + DWI + DCE-MRI
Two radiologists were asked to read each combination of parameters and rate them for the presence of cancer on a 5 point scale (1 = definitely absent to 5 = definitely present). Their evaluations were also compared with a clinical model (predictors such as PSA, Gleason score, tumor volume, tumor stage, etc.) and predictive accuracies were statistically calculated.
The team found that T2-weighted imaging + DWI performed significantly better than T2-weighted imaging alone. The combination of T2 + DWI + DCE gave no more prediction advantage than T2 + DWI. In addition, adding the T2 + DWI model to the clinical model outperformed the clinical model alone in predicting downgrading.
The authors concluded that “mpMRI improves the ability to identify a subgroup of patients with Gleason 3+4 PCa on biopsy who are candidates for active surveillance.” Thus, evaluation by mpMRI could help delay or avoid overtreatment of prostate cancer patients who may instead be eligible for active surveillance with image monitoring.
IMPORTANT NOTE: At the Sperling Prostate Center, we strongly believe that patients with a focus of Gleason 3+4 or 4+3 disease are not candidates for active surveillance. We recommend that those with a Gleason 7 tumor that is amenable to ablation undergo focal therapy. Active surveillance was only an alternative to radical treatment before focal ablation was established. Today, our treatment approach of focal laser ablation offers clinical control of Gleason 7 prostate cancer with little to no side effects.
[i] Sarici H, Telli O, Yigitbasi O, Ekici M et al. Predictors of Gleason score upgrading in patients with prostate biopsy Gleason score ?6. Can Urol Assoc J. 2014 May;8(5-6):E342-6. doi: 10.5489/cuaj.1499.
[ii] Gondo T, Poon B, Matsumoto K, Bernstein M, Sjoberg D, Eastham J. Clinical role of pathological downgrading after radical prostatectomy in patients with biopsy confirmed Gleason score 3 + 4 prostate cancer. BJUI Int. Aug 2014. [Epub ahead of print.] DOI:10.1111/bju.12769.
[iii] Gondo T, Hricak H, Sala E, Zheng J et al. Multiparametric 3T MRI for the prediction of pathological downgrading after radical prostatectomy in patients with biopsy-proven Gleason score 3?+?4 prostate cancer. Eur Radiol. 2014 Aug 7. [Epub ahead of print] DOI: 10.1007/s00330-014-3367-7.